Norbert Buchmuller [Tue, 25 Nov 2008 06:32:57 +0000]
* Wrapped SQL::Abstract::Test functionality in a new module (DBIC::SqlMakerTest).
* Made a test SKIP that does not work with SQL::Abstract < 1.49 (used to be a TODO test).
Norbert Buchmuller [Tue, 25 Nov 2008 04:58:38 +0000]
* Documented using PostgreSQL arrays in Cookbook.
Norbert Buchmuller [Tue, 25 Nov 2008 04:38:08 +0000]
* Fixed the test plan in 72pg.t.
* Fixed SQLA version check in SKIP tests (it's 1.49 in current 1.50RC branch).
Norbert Buchmuller [Tue, 25 Nov 2008 04:34:29 +0000]
* Added a test for arrayref bind values in search to match PostgreSQL array type values.
Norbert Buchmuller [Tue, 25 Nov 2008 04:23:16 +0000]
* Made some more tests SKIP if SQLA version < 1.50.
Norbert Buchmuller [Tue, 25 Nov 2008 04:13:10 +0000]
* Disabled stringification of arrayref bind values (used to pass values for PostgreSQL arrays).
Norbert Buchmuller [Tue, 25 Nov 2008 04:03:47 +0000]
* Made PostgreSQL array tests SKIP if SQLA version < 1.50.
Norbert Buchmuller [Fri, 21 Nov 2008 23:34:29 +0000]
* Added tests for passing arrayrefs as bind values for PostgreSQL array values.
Norbert Buchmuller [Fri, 21 Nov 2008 23:28:58 +0000]
* Updated svn:ignore.
Norbert Buchmuller [Fri, 21 Nov 2008 23:27:42 +0000]
* Added a WHATSNEW.txt file that summarises the changes on this branch.
Norbert Buchmuller [Fri, 21 Nov 2008 23:23:05 +0000]
* Documented order_by => [ { -desc => 'col' } ] syntax.
Norbert Buchmuller [Fri, 21 Nov 2008 23:11:16 +0000]
* Test cases for every supported order_by syntax.
Norbert Buchmuller [Mon, 17 Nov 2008 04:29:33 +0000]
* Added test cases to test if arrayref bind values in insert/update are passed through sql_maker intact.
* Added 'array_datatypes' parameter to the sql_maker constructor.
Norbert Buchmuller [Thu, 13 Nov 2008 06:22:53 +0000]
* Bumped SQL::Abstract version dependency.
Norbert Buchmuller [Thu, 13 Nov 2008 06:13:07 +0000]
* Made DBIC::SQL::Abstract pass on order_by hashref ({-desc => 'colname'}) to SQL::Abstract.
Matt S Trout [Wed, 12 Nov 2008 16:43:54 +0000]
Merge 'trunk' into 'sqla_1.50_compat'
r25478@agaton (orig r5092): ribasushi | 2008-11-10 22:46:45 +0000
Switch out one more reliance on connect_info
r25483@agaton (orig r5097): ribasushi | 2008-11-10 23:09:23 +0000
belongs_to relationships are explicitly flagged by default, to aid the DBIC SQLT parser (by abraxxa)
r25484@agaton (orig r5098): ribasushi | 2008-11-10 23:18:01 +0000
Clarify attributes argument of find()
r25499@agaton (orig r5113): plu | 2008-11-12 16:42:44 +0000
fixed 73oracle.t, added charfield
Johannes Plunien [Wed, 12 Nov 2008 16:42:44 +0000]
fixed 73oracle.t, added charfield
Peter Rabbitson [Mon, 10 Nov 2008 23:18:01 +0000]
Clarify attributes argument of find()
Peter Rabbitson [Mon, 10 Nov 2008 23:09:23 +0000]
belongs_to relationships are explicitly flagged by default, to aid the DBIC SQLT parser (by abraxxa)
Norbert Buchmuller [Tue, 11 Nov 2008 06:00:44 +0000]
* Removed dead code (sub _RowNumberOver).
Norbert Buchmuller [Tue, 11 Nov 2008 06:00:18 +0000]
* Removed a wrong TODO test: "order_by with quoting needs fixing (ash/castaway)" (asked by mst).
* Made a TODO test non-TODO: "select attr with star needs fixing (mst/nate)" (works with SQLA 1.50).
Norbert Buchmuller [Tue, 11 Nov 2008 05:59:32 +0000]
* Converted some of the test cases to use SQL::Abstract::Test.
* Added the parens for the join condition (SQL::Abstract::Test can't cope with it, and, besides, they help readability there; plus this gives the same output that older SQLA generated).
Peter Rabbitson [Mon, 10 Nov 2008 22:46:45 +0000]
Switch out one more reliance on connect_info
John Napiorkowski [Mon, 10 Nov 2008 16:12:28 +0000]
changed the way we get the DSN for debugging in replication
Peter Rabbitson [Sun, 9 Nov 2008 20:19:39 +0000]
Switch schema management version calls from VERSION() to schema_version()
Peter Rabbitson [Sun, 9 Nov 2008 16:02:05 +0000]
Don't rely on connect_info(), instead fetch the dsn from _dbi_connect_info (which is stable no matter which connect_info() flavor is used)
todo: ping jnpiorkowski about Replicated::Pool and Replicated::Replicant - it blindly assumes that connect_info->[0] contains the dsn, which might not be the case. Even when using _dbi_connect_info, the first element might be a coderef
Peter Rabbitson [Sun, 9 Nov 2008 15:35:30 +0000]
Fix detection of multiple rows being returned for single(): calling fetchrow_array in scalar context leads to indefined results (search.cpan.org/~timb/DBI/DBI.pm#fetchrow_array)
Peter Rabbitson [Sun, 9 Nov 2008 15:29:34 +0000]
Some testdb changes and alignment, preparing for test refactor branch
Peter Rabbitson [Sun, 9 Nov 2008 15:27:40 +0000]
SQLT behaves sanely in scalar context, no need for stupid version detections (ribasushi--)
Peter Rabbitson [Sun, 9 Nov 2008 13:35:57 +0000]
More TODO items
Peter Rabbitson [Fri, 7 Nov 2008 09:42:48 +0000]
One extra multicreate test and some cleanup
Peter Rabbitson [Fri, 7 Nov 2008 07:30:49 +0000]
Fix loss of objects on multicreate over several 1:M relations
Peter Rabbitson [Fri, 7 Nov 2008 07:29:21 +0000]
Final insanification of the multicreate test
Steffen Schwigon [Thu, 6 Nov 2008 21:59:42 +0000]
unhide SQL::Translator::Parser::DBIx::Class from pause indexer
Steffen Schwigon [Thu, 6 Nov 2008 20:23:22 +0000]
pod documentation for SQL::Translator::Parser::DBIx::Class
Michael G Schwern [Thu, 6 Nov 2008 12:44:36 +0000]
A little indentation clean up
Test the result of an eval directly.
discard the changes to $sandl to quiet an "object destroyed with changes"
warning.
Michael G Schwern [Thu, 6 Nov 2008 12:44:27 +0000]
mk_group_accessor() is sometimes called internally when inflating a column
and with an array ref field spec, but the accessor and column names are
the same. Special case that to go through the usual accessor_name_for stuff.
Michael G Schwern [Tue, 4 Nov 2008 18:26:41 +0000]
[rt.cpan.org 36863]
Fix mk_group_accessors() to handle [$field, $name]. It ignores
accessor/mutator_name_for() because if you pass in your own accessor
name you probably mean it.
Michael G Schwern [Tue, 4 Nov 2008 18:26:29 +0000]
Normalize indentation
Michael G Schwern [Tue, 4 Nov 2008 18:26:19 +0000]
And now really untabify
Michael G Schwern [Tue, 4 Nov 2008 18:26:09 +0000]
normalize the indentation and detabify
Johannes Plunien [Tue, 4 Nov 2008 16:22:51 +0000]
fixed 73oracle.t
Peter Rabbitson [Sun, 2 Nov 2008 16:55:31 +0000]
Test prompted by discussion about tasty stuff
Peter Rabbitson [Sun, 2 Nov 2008 16:54:46 +0000]
Better linking in Relatiomship POD
Peter Rabbitson [Sun, 2 Nov 2008 13:32:26 +0000]
More svn:ignore
Peter Rabbitson [Sun, 2 Nov 2008 13:30:49 +0000]
Clarify Relationship POD
Peter Rabbitson [Sun, 2 Nov 2008 12:28:02 +0000]
refactor of t/96multi_create.t:
- simplify test case submitted by wreis
- incorporate it into the large multicreate at the end
- eval{} independent sections of the test so more output can be gathered
Peter Rabbitson [Sat, 1 Nov 2008 20:58:07 +0000]
Augment commit r5037 - the column_info feature on SQLite is not buggy but partially unimplemented
Johannes Plunien [Sun, 2 Nov 2008 01:59:49 +0000]
rolling back my last change, waiting until new sqlt is on cpan
Johannes Plunien [Sun, 2 Nov 2008 01:37:21 +0000]
fixed next warnings
Johannes Plunien [Sun, 2 Nov 2008 01:22:09 +0000]
updated Changes
Johannes Plunien [Sun, 2 Nov 2008 01:19:09 +0000]
split sql statements for deploy only if SQLT::Producer returned a scalar
Peter Rabbitson [Sat, 1 Nov 2008 14:23:15 +0000]
TODOify a test broken by design, which exposes a bug in virtually all DBD::SQLite versions
Wallace Reis [Sat, 1 Nov 2008 00:42:56 +0000]
added failing test case for multi create
Peter Rabbitson [Fri, 31 Oct 2008 23:36:02 +0000]
Kill the install process at the MakeMaker stage if a dodgy DBD::SQLite is detected
Peter Rabbitson [Fri, 31 Oct 2008 20:34:27 +0000]
68inflate_has_a.t tests the cdbi-compat shim and thus belongs in t/cdbi-t
bumped up the version of DBD::SQLite required for testing
Peter Rabbitson [Fri, 31 Oct 2008 11:22:38 +0000]
Adjust MANIFEST.SKIP
Peter Rabbitson [Fri, 31 Oct 2008 11:21:41 +0000]
Cleanup TODO, delete implemented VERSIONING.SKETCH
Peter Rabbitson [Fri, 31 Oct 2008 00:26:58 +0000]
It seems that some CPAN testers have a weird unreleased DBD::SQLite 1.14001 which return unexpected column info - fixing. Should take care of:
http://www.nntp.perl.org/group/perl.cpan.testers/2008/10/msg2525124.html
http://www.nntp.perl.org/group/perl.cpan.testers/2008/10/msg2525137.html
Peter Rabbitson [Thu, 30 Oct 2008 21:30:47 +0000]
Dev-Release 0.08099_05
Peter Rabbitson [Thu, 30 Oct 2008 21:26:00 +0000]
Final set of DBI docfixes
Jess Robinson [Thu, 30 Oct 2008 21:10:05 +0000]
Documentation fixups, still needs standardising though
Peter Rabbitson [Thu, 30 Oct 2008 15:26:36 +0000]
Fix some lapses in Storage::DBI pod
Peter Rabbitson [Thu, 30 Oct 2008 12:41:04 +0000]
add some things to castaway's rewrite
Peter Rabbitson [Thu, 30 Oct 2008 12:36:54 +0000]
Add new TODO items
Peter Rabbitson [Thu, 30 Oct 2008 12:26:05 +0000]
Test direct sql passed to ResultSetColumn
Peter Rabbitson [Thu, 30 Oct 2008 11:20:09 +0000]
Massive cleanup of Storage::DBI::connect_info, based on patches from Oleg Kostyk
Jess Robinson [Wed, 29 Oct 2008 21:09:49 +0000]
Standardising and improving Row's docs. Added cross linking etc.
Peter Rabbitson [Wed, 29 Oct 2008 11:18:44 +0000]
Minor chaining patch + test by Norbert Buchmuller (chaining needs a massive cleanup, delayed until next release)
Nigel Metheringham [Tue, 28 Oct 2008 13:37:02 +0000]
Yet more RH perf bug test tweaks
Nigel Metheringham [Tue, 28 Oct 2008 13:07:44 +0000]
Reworked RH perf test to be more targetted
Peter Rabbitson [Tue, 28 Oct 2008 12:32:52 +0000]
Fix an obscure bug in the DBI exception handler, clobbering the exception contents when the same $dbh is used outside of dbic code
Tom Hukins [Tue, 28 Oct 2008 11:11:20 +0000]
Skip tests if DateTime is not installed.
Nigel Metheringham [Tue, 28 Oct 2008 09:22:42 +0000]
Merge 'doc_mods' into 'trunk'
Merged in changes to StartupCheck, documentation and tests for Red Hat perl performance issues
Nigel Metheringham [Tue, 28 Oct 2008 09:21:33 +0000]
Merge 'doc_mods' into 'trunk'
Merged in documentation modifications for Intro and Example to use load_namespaces syntax
Nigel Metheringham [Tue, 28 Oct 2008 09:08:29 +0000]
Missing semicolon
Nigel Metheringham [Tue, 28 Oct 2008 09:00:31 +0000]
Minor RH perf bug test tweaks including internal documentation
Nigel Metheringham [Mon, 27 Oct 2008 21:55:55 +0000]
Minor tweaks to RH perf issue stuff following suggestions from ribasushi>
Nigel Metheringham [Mon, 27 Oct 2008 21:42:51 +0000]
Reworked information on RH perl performance issues
Peter Rabbitson [Mon, 27 Oct 2008 17:27:40 +0000]
dirty_column detection in set_column() should not depend on the (undocumented) return value of store_column()
Johannes Plunien [Sun, 26 Oct 2008 23:38:27 +0000]
updated last changes of InflateColumn::DateTime
Peter Rabbitson [Sun, 26 Oct 2008 17:32:11 +0000]
Improve IC::DT timezone checker POD, add note explaining code weirdness
Johannes Plunien [Sun, 26 Oct 2008 21:32:29 +0000]
typo in pod
Johannes Plunien [Sun, 26 Oct 2008 21:31:17 +0000]
referencing DateTime/Floating_DateTimes
Johannes Plunien [Sun, 26 Oct 2008 21:24:18 +0000]
using local on SIG handler
Johannes Plunien [Sun, 26 Oct 2008 20:57:28 +0000]
floating timezone warning
Nigel Metheringham [Sun, 26 Oct 2008 12:47:10 +0000]
Reworked Manual::Intro to use load_namecpaces
Nigel Metheringham [Sun, 26 Oct 2008 12:13:17 +0000]
Updated example to use load_namespace
Nigel Metheringham [Sun, 26 Oct 2008 11:59:35 +0000]
Modified example to use load_namespace with default naming rather than load_classes
Johannes Plunien [Sat, 25 Oct 2008 03:11:43 +0000]
rafl refactored last changes
Johannes Plunien [Sat, 25 Oct 2008 02:39:36 +0000]
added myself to CONTRIBUTORS
Johannes Plunien [Sat, 25 Oct 2008 02:33:59 +0000]
possible to explicitly skip inflation, again rafl++
Johannes Plunien [Sat, 25 Oct 2008 02:05:00 +0000]
added new option to enforce Date or DateTime inflation, no matter what type the column is, rafl++ for having this idea
Peter Rabbitson [Fri, 24 Oct 2008 14:32:17 +0000]
complete a test
Peter Rabbitson [Fri, 24 Oct 2008 13:29:08 +0000]
Massive multi-create testing, with two regression fixes:
- An uninserted row must be passed to find_or_create_related with all uninserted objects on which it depends. Since currently we store those in {_inflated_column} as opposed to {_relationship_data} as we should, a simple ->get_inflated_columns will suffice
- We must find_or_create related rows, as find_or_new might queue multiple identical objects for insertions, which will subsequently violate existing unique constraints
Peter Rabbitson [Fri, 24 Oct 2008 13:23:33 +0000]
More test schema relations for future tests
Peter Rabbitson [Fri, 24 Oct 2008 13:21:40 +0000]
Count warnings differently, as SQLT produces some weird stuff under 5.10. All of this will be fixed after the test refactor
Michael G Schwern [Fri, 24 Oct 2008 03:00:57 +0000]
Annoying DBD::SQLite warning is tripping up Test::NoWarnings.
Work around it.
Michael G Schwern [Fri, 24 Oct 2008 02:54:25 +0000]
Using DBIx::Class as a parent causes Class::Accessor::Grouped to be
introduced to the hierarchy too early and thus CDBICompat::AccessorMapping's
own mk_group_accessor() not get called.
The only reason we're inheriting from DBIx::Class is to get mk_classdata().
Get it from Class::Data::Inheritable instead. mk_classdata() should be
moved out of DBIx::Class into a role to avoid this problem.
Nigel Metheringham [Thu, 23 Oct 2008 15:00:06 +0000]
Added warnning to $rs->delete which fires if delete is done on a joined/chained resultset